LAUSR

512 Background: Advances in prognostic allocation of pts with PSCC receiving regional LND are needed, along with an assessment of the role of perioperative treatments. Methods: An international, multicenter, retrospective database was queried for all pts with PSCC who had received inguinal ± pelvic LND from 1980 to 2017. We classified or re-classified pts according to the 2009 TNM classification system. The primary endpoint was OS. Survival analysis methods were adopted for the purpose of detecting and modeling putative prognostic factors, and develop a prognostic stratification tool. The analysis was performed in two steps: first, we modeled outcome data and covariates by resorting to a random forest (RF) method (an “ensemble” machine-learning approach, with an ad hoc approach incorporating missing data imputation). Second, data was modeled using Cox proportional hazard regression. In addition, a Cox model was fit including pre-specified variables that were deemed to be clinically relevant. Results: There were 743 pts who received LND at seven referral centers from Europe, the USA, Brazil, and China. Of these pts, 689 were analyzed: 86 (12.5%) received neoadjuvant chemotherapy (NAC); 171 (24.8%) received adjuvant chemotherapy (AC), and 74 (10.7%) received adjuvant radiotherapy. The variables significantly associated with OS were age (p < 0.001), the pathologically involved/total removed LN ratio (p < 0.001), pN stage (overall p < 0.001) and NAC (HR: 1.58, 95%CI: 1.10-2.26, p = 0.013). NAC and AC were ineffective or detrimental in N1-2 patients, whereas they provided OS improvements in N3 patients. OS improvement reached statistical significance for AC in pelvic LN-positive pts (p = 0.046). Finally, we developed a nomogram predicting 12- and 24-month OS based on pre-specified variables (c-index: 0.75). The study is limited by its retrospective nature. Conclusions: We propose a post-treatment prognostic tool that can be offered as an aid to physicians to enhance decision-making, clinical research, and patient counseling after LND in PSCC. Administration of NAC and AC should be restricted to N3 patients.